libcaps thread testing code committed

Matthew Dillon dillon at apollo.backplane.com
Mon Dec 8 10:17:44 PST 2003


:So its all synchronous at the moment?  Does this mean there are no
:queues or that, as in cooking, all the prep is done and we just need the
:proper recipe to combine the ingredients to bake up some asynchronous
:support.
:
:>     This commit 'proves out' the upcall API fairly well.  I was able to
:>     implement all the infrastructure necessary to support user 
:> threading,
:>     including IPI's between virtual cpus (rfork'd processes), without
:>     having to drop into the kernel to performing a thread switch.
:
:Groovy.

    Note quite synchronous... Galen has come up with a 'systhread' idea.
    The idea is that we will use the virtual cpu mechanism (which basically
    just does an rfork) to 'simulate' asynch system call messages.  This
    way the userland threading system can be written to use asynch system
    call messagse before actual kernel support is in place.

    The get_mplock() and rel_mplock() is the userland equivalent of a 
    big giant lock which is how we will protect things like malloc(),
    at least before we get our slab allocator tied in.

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>





More information about the Kernel mailing list